W. R. Berkley: Where should underwriting capacity go next?
Download case PDF (4 pages)Request teacher’s pack
Student case
Information available through January 26, 2026
In the first weeks of 2026, W. R. Berkley Corporation (Berkley) faced a question that did not appear in its group-wide premium growth number: where should the next dollar of underwriting capacity go?
Berkley had just reported record annual pre-tax underwriting income and strong investment income. Yet its January 1 property-catastrophe reinsurance renewal had repriced sharply, while its chief executive saw some insurers turning toward casualty to replace property premium growth. The shift created an opportunity and a risk. Specialty insurers can choose which policies and layers to write, but the loss cost on long-tail liability business may take years to emerge.
Assume you are advising a hypothetical Berkley operating committee responsible for setting 2026 underwriting appetite across business units. The committee has asked for a recommendation on where to pursue incremental growth, where to hold capacity steady, and where to decline business unless price or terms improve. You have no access to internal line-level loss picks, renewal submissions, or capital models. Your job is to make a disciplined recommendation from public evidence and state what information you would request before implementation.
A specialty insurer with many choices
Berkley is a property-casualty insurance holding company with two reporting segments: Insurance and Reinsurance & Monoline Excess. Its Insurance businesses write commercial and specialty personal insurance, including admitted and excess-and-surplus (E&S) business, in the United States and other markets. Reinsurance & Monoline Excess writes treaty and facultative reinsurance and retains certain risks on an excess basis. The company’s operating units specialize in different industries, customer groups and coverages.
This structure provides flexibility. Berkley can direct appetite toward different types of insureds, coverages, attachment points and geographies. It also makes group totals hard to interpret: one fast-growing line can hide a contracting line, and a low combined ratio in one segment cannot tell management whether the next account is correctly priced.
At year-end 2024, Insurance generated $10.55 billion of net premiums written, compared with $1.42 billion in Reinsurance & Monoline Excess. During 2025, Berkley’s consolidated net premiums written rose 6.2% to $12.71 billion. Reported full-year operating return on equity was 20.6%. Fourth-quarter pre-tax underwriting income reached $338 million, and the calendar-year combined ratio was 89.4%. The company also reported an 87.9% current accident-year combined ratio before catastrophe losses.
These results gave the group resources and confidence. They did not reveal the expected return on a marginal property treaty or a new casualty policy. Reported profitability also combines current claims, catastrophe experience, prior-year reserve development, operating expenses, and the timing of earned premiums.
The January renewal sends a different signal
On January 26, CEO W. Robert Berkley Jr. described an increasingly competitive property market. He said the risk-adjusted rate on the company’s main property-catastrophe treaty had fallen 19% at the January 1 renewal. He characterized large-account, layered property as especially competitive.
The 19% figure referred to the main property-cat treaty. It was not a company-wide property rate decline, and the call did not provide transaction details that would separate rate from changes in exposure, attachment, limits or other contract terms. In the same quarter, Berkley reported average rate increases of approximately 7.1%, excluding workers’ compensation, across its broader business. The figures describe different parts of the portfolio and different measures.
The CEO also said that some insurers, having difficulty reaching premium targets in property, appeared to be leaning into casualty. He described that as an early sign of spillover. This was management’s view of competitors’ behavior, not a verified explanation for all market pricing. He noted that property-cat had rebounded earlier in the cycle, while casualty had not experienced a similar bounce.
Uneven prospects within casualty
Berkley did not describe casualty as uniformly attractive. Management said auto liability remained difficult. D&O and architects-and-engineers professional liability were also challenging. By contrast, management pointed to possible opportunity in smaller casualty accounts, excess and umbrella coverage, E&S, the standard market, and medical stop loss.
Even within a line, the risk varies. A primary layer may respond frequently to smaller claims; an excess layer may attach only after substantial losses but can accumulate severe claims across insureds. A commercial auto book depends on fleet, geography, limits, litigation and driver characteristics. Medical stop-loss covers employer health plans above specified attachment points, exposing the insurer to different drivers than general liability. E&S business can address risks that do not fit standard forms, but individualized terms do not remove uncertainty.
Management described a line-by-line process: estimate the loss pick, compare it with the rate available, judge confidence in the estimate, and consider how long claims may take to develop. Some business may justify more growth; elsewhere, Berkley may decline an account if the price it needs is unavailable. The company did not disclose these internal loss picks, thresholds or line-specific budgets.
The timing problem
Property-catastrophe claims can be severe and volatile, while liability claims may be reported or settled over extended periods. A price that appears profitable based on early experience may be inadequate if claim frequency, severity, legal outcomes or medical costs develop differently than expected. Conversely, refusing all growth in a softening market may surrender relationships and attractive risks to competitors.
Academic research offers more than one explanation for changing insurance prices. Capacity constraints can make the supply of coverage tighten after shocks, while competition and pricing practices can also produce premiums that do not keep pace with future claims. Research on historical general-liability markets highlights the difficulty of predicting long-tail losses and the possibility that competition intensifies before reserve problems become visible. But these studies do not predict Berkley’s 2026 loss costs. They support asking what is known about each risk, how quickly experience will emerge, and how much confidence management should place in its estimate.
An external reference reinforces the importance of portfolio specificity. Travelers reported Q4 Business Insurance renewal premium change of 6.1%, 85% retention and $675 million of new business. Select Accounts and Middle Market net written premiums grew 4% and 3%, respectively, while National Property declined 3%, which Travelers attributed to disciplined underwriting. Its release defines renewal premium change as including both rate and exposure changes; renewal rate change excludes exposure. Travelers’ mix differs from Berkley’s, so these figures are not like-for-like rates, but they show how growth and contraction can coexist across commercial books.
The committee’s decision
The hypothetical operating committee must set an appetite framework for 2026. It can maintain discipline and protect capital by declining more business where price is inadequate; selectively pursue lines where Berkley believes its expertise supports a sound loss pick; or adopt a broader measured-growth plan, with caps and frequent reviews to preserve distribution and market position.
The decision must account for the gap between observable rate changes and unknown expected losses. The company’s positive overall results may support investment in underwriting capacity, but group-level profitability is not a substitute for line-level economics. A proposed growth area must have a credible reason that Berkley can select or price the risk better, not merely evidence that premium is available.
Discussion questions
- What does the 19% risk-adjusted decrease in Berkley’s main property-cat treaty establish? What does it not establish about the rest of the company’s portfolio?
- Which two or three businesses would you prioritize for incremental underwriting capacity, and which would you constrain? Explain how expected loss, achievable price, exposure accumulation, and claim-development time affect your choices.
- Should Berkley respond to competitors leaning into casualty? What evidence would distinguish attractive growth from premium chasing?
- What line-level information would you request before setting budgets? Identify early warning indicators and rules for changing appetite during 2026.
- How would you balance near-term underwriting margin, long-term broker/customer relationships, and the possibility that market conditions reverse?
Selected evidence
| Measure / statement | Public evidence at cutoff | Interpretation caution |
|---|---|---|
| 2025 NPW | $12.711bn vs $11.972bn in 2024 | Group growth is not a line-level profitability measure |
| 2025 operating ROE | 20.6% | Company-defined operating metric, based on beginning equity |
| Q4 underwriting | $338m pre-tax underwriting income; 89.4% calendar-year combined ratio; 87.9% current accident-year ex-cat ratio | Different measures and time bases; current ratio excludes catastrophe losses |
| Q4 average rate increase | About 7.1%, excluding workers’ compensation | Broad portfolio average; not a property-cat treaty rate |
| Jan. 1 main property-cat treaty | Risk-adjusted rate down 19% | Management-reported, treaty-specific renewal; contract detail not disclosed |
| Management market view | Casualty competition may be increasing; smaller casualty, umbrella, E&S and medical stop loss seen as brighter spots | Management interpretation and opportunity claims, not verified margins |
Sources: W. R. Berkley Q4/FY2025 results release (Jan. 26, 2026), company IR; W. R. Berkley FQ4 2025 earnings call transcript, PDF hosted by company IR; W. R. Berkley 2024 Form 10-K, SEC; Travelers Q4/FY2025 results (Jan. 21, 2026), company IR. Academic context: Weiss (2007); Harrington and Danzon (1994). Full source details and reading limits are in the accompanying research dossier.
This case is fictionalized only in the committee assignment. It contains no invented private meeting, quote, management decision or confidential company data. It uses information available through January 26, 2026; subsequent results and developments are excluded.